This article intends to dive into how the need for DevOps culture has evolved in the software industry over the recent years and how various tools, processes, and technologies have helped bring value to a lot of teams across the world that build amazing software at a very high pace.
In the last decade we have seen a lot of evident changes in how we as end users interact with software applications. In practice, it has been noted that most applications were developed for only a few target platforms — mostly legacy applications that were deployed in the customer’s infrastructure.
The software we use today exists in multiple different forms and platforms and this has been transitioning to become the industry standard — Native Apps(Windows, iOS, macOS, Android, Linux), WebApps(SaaS, Progressive Web Apps)
With such a high rate of evolution in this industry, we also need to have processes and tools that support this kind of velocity and help organizations and teams build, test, and deploy applications and features faster than ever before — this is where DevOps and Cloud bring value.
DevOps Practices that bring Value to Teams
Automated Pipelines — CI/CD
Pipelines are where developed application code is built, tested and packaged into deployable artifacts, and finally deployed — this is analogous to the logistics involved in moving manufactured products from factories to the consumers. If this is a slow and disorganized process, the products would be manufactured on time but will reach customers with a delay — This is why automation is of extreme importance in building and delivering software. Automation will ensure even the most minor changes will reach end-users at a high pace.
Timing is of utmost importance to bring maximum value — features should be built and deployed as soon as the requirements arise.
Infrastructure as Code — IaC
Gone are the days when server and other infrastructure provisioning was a manual task, today with tools like Terraform, Azure ARM Templates, AWS CloudFormation Templates we can write infrastructure definition as code, and the cloud platform will provision infra interpreting the code definitions from these templates with high consistency. This will ensure that -
- Infrastructure provisioning is repeatable and not time-consuming
- Infra definition can be peer-reviewed with Pull Requests just like how application code is peer-reviewed
- Helps automate deployment and reduces the risk of human error
- Almost 0 environment parity — Development and Test Environments identical to the production
Configuration Management tools like Ansible, Chef, Puppet, and DSC help in managing the consistent state of servers where the application services are deployed to. This will ensure that -
- All environments are of the same state — the runtime versions, web server versions, are containerization platform versions, etc.
- Configuration drifts are automatically handled
- The application code will work exactly the same way in all environments
Also since configuration management is written as code, the changes to server state can be tracked in code version control systems like Git.
As the application is constantly growing with new features and tons of small changes and deployments, we need to have a way to monitor how these continuous updates are impacting the end-user experience — here’s where APM tools and Infrastructure monitoring tools play an important role.
They capture data based on certain defined metrics and provide more insights on this captured data. This will help the development team get a better understanding of how the end-users are actually using the application, where the performance anomalies are, thus helping the team troubleshoot and get to the root cause of issues before they are reported by the customers themselves.
Future of DevOps Adoption
Considering the pace at which organizations are adopting DevOps and Cloud technologies, the only concern at this time is — if the technology decisions related to these powerful tools, processes, and technologies are not made wisely, it could potentially have a negative impact on the overall software development project — increased cost and complexity, which is why it is important to transition to the DevOps culture with care and expertise.
Implementation of the right DevOps stack taking careful consideration of the use case and requirements of the project and also ensuring that there is learning & cultural transition occurring within the team will enable the organization to leverage the maximum value in the form of — improved team collaboration, shorter release cycles, faster deployments, higher scalability, continuous improvement and finally the most important — customer delight.
Author: Varun Lokeswaran